Nodes from the Underground: Causal and Probabilistic Approaches for Complex Transportation Networks

An efficient transportation system is vital to the economic and social well-being of large cities. The transport demand implied by economic growth, however, requires transport networks to become more and more complex, making their management difficult. Fortunately, modern systems such as the London Underground generate vast amounts of data that can be analysed to better understand passenger behaviour and needs. Besides understanding the typical daily patterns that we can observe on a regular basis, Data Science methods allows us to look into in the less usual events such as unplanned disruptions that are still important to any user, and to also model individualised behaviour instead of only aggregates.

In a large system such as the London Underground, signal failures and disruptive events eventually take place, requiring passengers to change plans in a variety of ways. This research provides advanced statistical modelling and machine learning approaches to learn from past events to examine how passengers adapt themselves when a disruption occurs. When a disruption takes place, the model will provide information of likely changes, such as increased number of passengers leaving a station because they could not reach their destination. These models are important for transport authorities to understand the resilience of the system, different combinations of location and time of a disruption, and unusual responses from passengers that may motivate different communication strategies to inform users of better travel adjustments. This research also opens up conceptual ideas to be exploited in the future using new technologies to monitor and adaptively respond to passenger needs in a more optimised and time-effective way.

Among the potential non-academic beneficiaries, we list:

1. Transport authorities such as Transport for London. Although most of the transport authorities have their own analytics teams and vast expertise, they are welcoming of academic research, as made explicit in the Letter of Support provided to us by TfL. In particular, research motivated by basic methodology is important to them, as it focuses on complex models which are not straightforward to integrate with current systems, but which in the long run offers offer potentially important contributions. In addition, by the completion of our research, we will provide these beneficiaries with methodologies and tools that will enhance the operational management of public urban railway systems.

2. Users of large transport systems. Millions of daily users of large transport systems such as London Underground have the potential to benefit directly from a system that reacts quickly and adaptively to disruptions, saving time and reducing the stress of commuting in a big city. This includes secondary effects on productivity gained by more efficient transportation in the society as a whole. In January 2015, Dr Silva and Dr Kang presented preliminary work related to this proposal at the Science Museum "Lates" event, in South Kensington. There was an overwhelmingly positive response from the museum attendees to the type of work we intend to continue in depth with this grant. The general public, and Underground users among them, appreciates the fact that scientific community engages in rigorous research to improve their quality of life.

3. City administrators. Good urban infrastructure has long-term economic benefits. The confidence given by a state-of-the-art transport system is a major factor boosting the profile of cities that expect to be labelled as "world-class" environments for businesses and quality of life. Much of London's reputation, for instance, can be traced back to reliance on its public transport network, including the oldest underground network in the world. Knowing the limitations of its transport system as facilitated by new technologies provides crucial information for the city management.

4. Companies working on complex systems and smart city solutions. Understanding behaviour under disruption is relevant to organisations such as taxi companies, which may benefit for the outcomes of this research as knowing excess demand generated from passengers interrupting their journey can also be used to allocate more vehicles to serve customers that would otherwise arrive late at their destinations. Companies who provide general apps of interest for city dwellers may potentially benefit from APIs streaming forecasts of reactions to disruptions in the transportation network, as some users may be interested in avoiding overcrowded locations. We intend to work closely with these beneficiaries, for example, through UCL's Knowledge Transfer resources.

5. Other companies/researchers working on transport problems outside the domain of passenger movement. The ideas developed in this project are also relevant to other transportation problems in networks, such as computer networks and other applications discussed in research areas such as network tomography. Part of the technology on computational methods developed here can be transferred to these domains mainly through general multi-disciplinary journals as well as journals and conferences in targeted fields.